NetCDF Territorial Classifications and Python Colab Scripts Based on Low-Pass Filtered SbAI Time-Series Signals for Periods Longer Than 2 Years in Southern South America (Argentina, Chile, and Uruguay), Including StandardScaler-Normalized and Non-Normalized Versions. (TS-LP2 and TS-LP2S)

Abstract

This notebook implements a complete pipeline for FIR lowpass filtering and spatial classification of SbAI data:

- Applies a FIR lowpass filter (>2 years, 101 taps) to isolate long‑term variability. - Normalizes and interpolates missing values along the time axis. - Saves filtered series as NetCDF for reproducibility. - Performs clustering with KMeans++ using two scaling strategies: - RobustScaler (TS-LP2). - StandardScaler with transpose (TS-LP2S). - Generates classification maps with geographic boundaries. - Extracts cluster centroids and plots representative time series. - Computes clustering quality metrics (Silhouette, explained variance, relative reconstruction error).